Class: DataDuck::Table

Inherits:
Object
  • Object
show all
Defined in:
lib/dataduck/table.rb

Direct Known Subclasses

IntegrationTable

Class Attribute Summary collapse

Instance Attribute Summary collapse

Class Method Summary collapse

Instance Method Summary collapse

Class Attribute Details

.actions ⇒ Object

Returns the value of attribute actions.


8
9
10
 
# File 'lib/dataduck/table.rb', line 8

def actions
  @actions
end

.output_schema ⇒ Object

Returns the value of attribute output_schema.


7
8
9
 
# File 'lib/dataduck/table.rb', line 7

def output_schema
  @output_schema
end

.sources ⇒ Object

Returns the value of attribute sources.


6
7
8
 
# File 'lib/dataduck/table.rb', line 6

def sources
  @sources
end

Instance Attribute Details

#data ⇒ Object

Returns the value of attribute data.


11
12
13
 
# File 'lib/dataduck/table.rb', line 11

def data
  @data
end

#errors ⇒ Object

Returns the value of attribute errors.


12
13
14
 
# File 'lib/dataduck/table.rb', line 12

def errors
  @errors
end

Class Method Details

.output(schema) ⇒ Object 


42
43
44
45
 
# File 'lib/dataduck/table.rb', line 42

def self.output(schema)
  self.output_schema ||= {}
  self.output_schema.merge!(schema)
end

.source(source_name, source_table_or_query = nil, source_columns = nil) ⇒ Object 


26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
 
# File 'lib/dataduck/table.rb', line 26

def self.source(source_name, source_table_or_query = nil, source_columns = nil)
  self.sources ||= []

  source_spec = {}
  if source_table_or_query.respond_to?(:to_s) && source_table_or_query.to_s.downcase.include?('select ')
    source_spec = {query: source_table_or_query}
  elsif source_columns.nil? && source_table_or_query.respond_to?(:each)
    source_spec = {columns: source_table_or_query, table_name: DataDuck::Util.camelcase_to_underscore(self.name)}
  else
    source_spec = {columns: source_columns, table_name: source_table_or_query.to_s}
  end

  source_spec[:source] = DataDuck::Source.source(source_name)
  self.sources << source_spec
end

.transforms(transformation_name) ⇒ Object 


14
15
16
17
 
# File 'lib/dataduck/table.rb', line 14

def self.transforms(transformation_name)
  self.actions ||= []
  self.actions << [:transform, transformation_name]
end

.validates(validation_name) ⇒ Object 


20
21
22
23
 
# File 'lib/dataduck/table.rb', line 20

def self.validates(validation_name)
  self.actions ||= []
  self.actions << [:validate, validation_name]
end

Instance Method Details

#actions ⇒ Object 


47
48
49
50
51
52
53
54
55
56
 
# File 'lib/dataduck/table.rb', line 47

def actions
  my_actions = []
  for_class = self.class
  while for_class < Table
    my_actions.concat(for_class.actions || [])
    for_class = for_class.superclass
  end

  my_actions
end

#autogenerate_identity? ⇒ Boolean

Returns:

  • (Boolean) 

222
223
224
 
# File 'lib/dataduck/table.rb', line 222

def autogenerate_identity?
  false
end

#batch_size ⇒ Object 


202
203
204
 
# File 'lib/dataduck/table.rb', line 202

def batch_size
  nil
end

#building_name ⇒ Object 


226
227
228
 
# File 'lib/dataduck/table.rb', line 226

def building_name
  self.should_fully_reload? ? self.staging_name : self.name
end

#check_table_valid! ⇒ Object 


58
59
60
61
62
63
 
# File 'lib/dataduck/table.rb', line 58

def check_table_valid!
  if !self.batch_size.nil?
    raise "Table #{ self.name }'s batch_size must be > 0" unless self.batch_size > 0
    raise "Table #{ self.name } has batch_size defined but no extract_by_column" if self.extract_by_column.nil?
  end
end

#create_column_names ⇒ Object 


242
243
244
 
# File 'lib/dataduck/table.rb', line 242

def create_column_names
  self.create_schema.keys.map(&:to_s).sort
end

#create_schema ⇒ Object 


234
235
236
237
238
239
240
 
# File 'lib/dataduck/table.rb', line 234

def create_schema
  if self.autogenerate_identity?
    Util.deep_merge(output_schema, {dataduck_identity: 'bigint identity(1, 1)'}) # Redshift only
  else
    output_schema
  end
end

#distribution_key ⇒ Object 


65
66
67
68
69
70
71
 
# File 'lib/dataduck/table.rb', line 65

def distribution_key
  if self.output_column_names.include?("id")
    "id"
  else
    nil
  end
end

#distribution_style ⇒ Object 


73
74
75
 
# File 'lib/dataduck/table.rb', line 73

def distribution_style
  nil
end

#etl!(destinations, options = {}) ⇒ Object 


77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
 
# File 'lib/dataduck/table.rb', line 77

def etl!(destinations, options = {})
  if destinations.length != 1
    raise ArgumentError.new("DataDuck can only etl to one destination at a time for now.")
  end

  if options[:dates].nil?
    options[:dates] = [Date.today]
  end

  self.check_table_valid!

  destination = destinations.first

  if self.should_fully_reload?
    destination.drop_staging_table!(self)
  end

  data_processed = false
  batch_number = 0
  while batch_number < 1_000
    batch_number += 1
    self.extract!(destination, options)
    if self.data.length > 0
      self.transform!
      self.data.compact!
      self.load!(destination) if self.data.length > 0
      data_processed = true
    end

    if self.batch_size.nil?
      break
    else
      if self.batch_size == self.data.length
        DataDuck::Logs.info "Finished batch #{ batch_number }, continuing with the next batch"
      else
        DataDuck::Logs.info "Finished batch #{ batch_number } (last batch)"
        break
      end
    end
  end

  self.data = []

  if data_processed
    if self.should_fully_reload?
      destination.finish_fully_reloading_table!(self)
    end

    self.postprocess!(destination, options)
  else
    DataDuck::Logs.info "No data extracted for table #{ self.name }"
  end
end

#extract!(destination = nil, options = {}) ⇒ Object 


131
132
133
134
135
136
137
138
139
140
141
142
143
 
# File 'lib/dataduck/table.rb', line 131

def extract!(destination = nil, options = {})
  DataDuck::Logs.info "Extracting table #{ self.name }"

  self.errors ||= []
  self.data = []
  self.class.sources.each do |source_spec|
    source = source_spec[:source]
    my_query = self.extract_query(source_spec, destination)
    results = source.query(my_query)
    self.data.concat(results)
  end
  self.data
end

#extract_by_clause(value) ⇒ Object 


171
172
173
174
175
176
177
 
# File 'lib/dataduck/table.rb', line 171

def extract_by_clause(value)
  if value
    "WHERE #{ self.extract_by_column } >= '#{ value }'"
  else
    ""
  end
end

#extract_by_column ⇒ Object 


206
207
208
209
210
 
# File 'lib/dataduck/table.rb', line 206

def extract_by_column
  return 'updated_at' if self.output_column_names.include?("updated_at")

  nil
end

#extract_query(source_spec, destination = nil) ⇒ Object 


145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
 
# File 'lib/dataduck/table.rb', line 145

def extract_query(source_spec, destination = nil)
  escape_char = source_spec[:source].escape_char

  base_query = source_spec.has_key?(:query) ? source_spec[:query] :
     "SELECT #{ escape_char }#{ source_spec[:columns].sort.join(escape_char + ',' + escape_char) }#{ escape_char } FROM #{ source_spec[:table_name] }"

  extract_part = ""
  limit_part = self.limit_clause

  if self.extract_by_column
    if destination.table_names.include?(self.building_name)
      extract_by_column_without_table = self.extract_by_column.include?(".") ? self.extract_by_column.split(".").last : self.extract_by_column
      extract_by_value = destination.query("SELECT MAX(#{ extract_by_column_without_table }) AS val FROM #{ self.building_name }").first
      extract_by_value = extract_by_value.nil? ? nil : extract_by_value[:val]

      extract_part = self.extract_by_clause(extract_by_value)
    end
  end

  if base_query.downcase.split("from").last.include?(' where ')
    extract_part.gsub!('WHERE ', 'AND ')
  end

  [base_query, extract_part, limit_part].join(' ').strip
end

#identify_by_columns ⇒ Object 


212
213
214
215
216
 
# File 'lib/dataduck/table.rb', line 212

def identify_by_columns
  return ["id"] if self.output_column_names.include?("id")

  []
end

#include_with_all? ⇒ Boolean

Returns:

  • (Boolean) 

191
192
193
 
# File 'lib/dataduck/table.rb', line 191

def include_with_all?
  true
end

#indexes ⇒ Object 


195
196
197
198
199
200
 
# File 'lib/dataduck/table.rb', line 195

def indexes
  which_columns = []
  which_columns << "id" if self.output_column_names.include?("id")
  which_columns << "created_at" if self.output_column_names.include?("created_at")
  which_columns
end

#limit_clause ⇒ Object 


179
180
181
182
183
184
185
 
# File 'lib/dataduck/table.rb', line 179

def limit_clause
  if self.extract_by_column && self.batch_size
    "ORDER BY #{ self.extract_by_column } LIMIT #{ self.batch_size }"
  else
    ""
  end
end

#load!(destination) ⇒ Object 


187
188
189
 
# File 'lib/dataduck/table.rb', line 187

def load!(destination)
  destination.load_table!(self)
end

#name ⇒ Object 


296
297
298
299
300
301
302
303
 
# File 'lib/dataduck/table.rb', line 296

def name
  fixed_name = DataDuck::Util.camelcase_to_underscore(self.class.name)
  if fixed_name.start_with?("data_duck/")
    fixed_name = fixed_name.split("/").last
  end

  self.prefix + fixed_name
end

#output_column_names ⇒ Object 


250
251
252
 
# File 'lib/dataduck/table.rb', line 250

def output_column_names
  self.output_schema.keys.map(&:to_s).sort
end

#output_schema ⇒ Object 


246
247
248
 
# File 'lib/dataduck/table.rb', line 246

def output_schema
  self.class.output_schema || self.class.superclass.output_schema || {}
end

#postprocess!(destination, options = {}) ⇒ Object 


254
255
256
 
# File 'lib/dataduck/table.rb', line 254

def postprocess!(destination, options = {})
  destination.postprocess!(self)
end

#prefix ⇒ Object 


305
306
307
 
# File 'lib/dataduck/table.rb', line 305

def prefix
  ""
end

#recreate!(destination) ⇒ Object 


258
259
260
 
# File 'lib/dataduck/table.rb', line 258

def recreate!(destination)
  destination.recreate_table!(self)
end

#should_fully_reload? ⇒ Boolean

Returns:

  • (Boolean) 

218
219
220
 
# File 'lib/dataduck/table.rb', line 218

def should_fully_reload?
  false
end

#show ⇒ Object 


262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
 
# File 'lib/dataduck/table.rb', line 262

def show
  puts "Table #{ self.name }"
  self.class.sources.each do |source_spec|
    puts "\nSources from #{ source_spec[:table_name] || source_spec[:query] } on #{ source_spec[:source].name }"
    source_spec[:columns].each do |col_name|
      puts "  #{ col_name }"
    end
  end

  puts "\nOutputs "
  num_separators = self.output_schema.keys.map { |key| key.length }.max
  self.output_schema.each_pair do |name, datatype|
    puts "  #{ name }#{ ' ' * (num_separators + 2 - name.length) }#{ datatype }"
  end
end

#staging_name ⇒ Object 


230
231
232
 
# File 'lib/dataduck/table.rb', line 230

def staging_name
  "zz_dataduck_#{ self.name }"
end

#transform! ⇒ Object 


278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
 
# File 'lib/dataduck/table.rb', line 278

def transform!
  DataDuck::Logs.info "Transforming table #{ self.name }"

  self.errors ||= []
  self.actions.each do |action|
    action_type = action[0]
    action_method_name = action[1]
    if action_type == :transform
      self.data.map! { |row| self.public_send(action_method_name, row) }
    elsif action_type == :validate
      self.data.each do |row|
        error = self.public_send(action_method_name, row)
        self.errors << error if !error.blank?
      end
    end
  end
end